Search before asking
Version
4.1.3 lastest
What's Wrong?
我的pgsql数据库使用了postgis插件,该插件引入了新的数据类型,其中我有一些表使用了geomtry这一几何类型字段,该字段在cdc同步时会先报错类型不支持,从而导致“exclude_columns”排除参数无法生效。
以下排除可以对不含geometry类型的pgsql库表生效
CREATE JOB pg_szt_cdc
ON STREAMING
FROM POSTGRES (
"jdbc_url" = "jdbc:postgresql://192.168.1.100:5432/test",
"driver_url" = "postgresql-42.7.x.jar",
"driver_class" = "org.postgresql.Driver",
"user" = "doris_cdc",
"password" = "你的密码",
"database" = "test",
"schema" = "public",
"include_tables" =
"person,people",
"table.person.exclude_columns" = "name",
"table.people.exclude_columns" = "name",
"offset" = "initial"
)
TO DATABASE szt_ods (
"table.create.properties.replication_num" = "1"
);
以下排除可以对含geometry类型的pgsql库表不生效,无法正常执行
CREATE JOB pg_szt_cdc
ON STREAMING
FROM POSTGRES (
"jdbc_url" = "jdbc:postgresql://192.168.1.100:5432/test",
"driver_url" = "postgresql-42.7.x.jar",
"driver_class" = "org.postgresql.Driver",
"user" = "doris_cdc",
"password" = "你的密码",
"database" = "test",
"schema" = "public",
"include_tables" =
"building",
"table.building.exclude_columns" = "geometry",
"offset" = "initial"
)
TO DATABASE szt_ods (
"table.create.properties.replication_num" = "1"
);
building表结构如下:
/*
Navicat Premium Data Transfer
Source Server : 192.168.1.100test
Source Server Type : PostgreSQL
Source Server Version : 160013 (160013)
Source Host : 192.168.1.100:5432
Source Catalog : tp_szt
Source Schema : public
Target Server Type : PostgreSQL
Target Server Version : 160013 (160013)
File Encoding : 65001
Date: 03/09/2026 17:06:05
*/
-- Table structure for building
DROP TABLE IF EXISTS "public"."building";
CREATE TABLE "public"."building" (
"id" int8 NOT NULL,
"buildingid" int8 NOT NULL,
"buildingdata" jsonb,
"buildingclass" int8,
"buildingname" varchar(255) COLLATE "pg_catalog"."default",
"geometry" geometry(GEOMETRYCOLLECTION, 4326),
"create_by" varchar(32) COLLATE "pg_catalog"."default",
"create_time" timestamp(6),
"update_by" varchar(32) COLLATE "pg_catalog"."default",
"update_time" timestamp(6),
"processdata" jsonb,
"lng" numeric(16,8),
"lat" numeric(16,8)
)
;
COMMENT ON COLUMN "public"."building"."buildingid" IS '3D建筑模型的ID';
COMMENT ON COLUMN "public"."building"."buildingdata" IS '3D建筑模型的geojosn数据';
COMMENT ON COLUMN "public"."building"."buildingclass" IS '类别';
COMMENT ON COLUMN "public"."building"."buildingname" IS '3D建筑模型的名称';
COMMENT ON COLUMN "public"."building"."geometry" IS '面数据';
COMMENT ON COLUMN "public"."building"."create_by" IS '创建人';
COMMENT ON COLUMN "public"."building"."create_time" IS '创建时间';
COMMENT ON COLUMN "public"."building"."update_by" IS '更新人';
COMMENT ON COLUMN "public"."building"."update_time" IS '更新时间';
COMMENT ON COLUMN "public"."building"."processdata" IS 'mvt数据';
-- Indexes structure for table building
CREATE INDEX "buildingid_building_index" ON "public"."building" USING btree (
"buildingid" "pg_catalog"."int8_ops" ASC NULLS LAST
);
CREATE INDEX "geometry_building_index" ON "public"."building" USING gist (
"geometry" "public"."gist_geometry_ops_2d"
);
CREATE INDEX "id_building_index" ON "public"."building" USING btree (
"id" "pg_catalog"."int8_ops" ASC NULLS LAST
);
CREATE INDEX "st_buffer_building_index" ON "public"."building" USING gist (
st_buffer(geometry, 0::double precision) "public"."gist_geometry_ops_2d"
);
-- Primary Key structure for table building
ALTER TABLE "public"."building" ADD CONSTRAINT "building_origin_copy1_pkey" PRIMARY KEY ("id");
What You Expected?
能否调整排除优先级,让exclude_columns先与类型识别生效?
或者
对不支持或无法识别的数据类型统一默认为string,而不是直接失败?
How to Reproduce?
No response
Anything Else?
No response
Are you willing to submit PR?
Code of Conduct
Search before asking
Version
4.1.3 lastest
What's Wrong?
我的pgsql数据库使用了postgis插件,该插件引入了新的数据类型,其中我有一些表使用了geomtry这一几何类型字段,该字段在cdc同步时会先报错类型不支持,从而导致“exclude_columns”排除参数无法生效。
以下排除可以对不含geometry类型的pgsql库表生效
CREATE JOB pg_szt_cdc
ON STREAMING
FROM POSTGRES (
"jdbc_url" = "jdbc:postgresql://192.168.1.100:5432/test",
"driver_url" = "postgresql-42.7.x.jar",
"driver_class" = "org.postgresql.Driver",
)
TO DATABASE szt_ods (
"table.create.properties.replication_num" = "1"
);
以下排除可以对含geometry类型的pgsql库表不生效,无法正常执行
CREATE JOB pg_szt_cdc
ON STREAMING
FROM POSTGRES (
"jdbc_url" = "jdbc:postgresql://192.168.1.100:5432/test",
"driver_url" = "postgresql-42.7.x.jar",
"driver_class" = "org.postgresql.Driver",
)
TO DATABASE szt_ods (
"table.create.properties.replication_num" = "1"
);
building表结构如下:
/*
Navicat Premium Data Transfer
Source Server : 192.168.1.100test
Source Server Type : PostgreSQL
Source Server Version : 160013 (160013)
Source Host : 192.168.1.100:5432
Source Catalog : tp_szt
Source Schema : public
Target Server Type : PostgreSQL
Target Server Version : 160013 (160013)
File Encoding : 65001
Date: 03/09/2026 17:06:05
*/
-- Table structure for building
DROP TABLE IF EXISTS "public"."building";
CREATE TABLE "public"."building" (
"id" int8 NOT NULL,
"buildingid" int8 NOT NULL,
"buildingdata" jsonb,
"buildingclass" int8,
"buildingname" varchar(255) COLLATE "pg_catalog"."default",
"geometry" geometry(GEOMETRYCOLLECTION, 4326),
"create_by" varchar(32) COLLATE "pg_catalog"."default",
"create_time" timestamp(6),
"update_by" varchar(32) COLLATE "pg_catalog"."default",
"update_time" timestamp(6),
"processdata" jsonb,
"lng" numeric(16,8),
"lat" numeric(16,8)
)
;
COMMENT ON COLUMN "public"."building"."buildingid" IS '3D建筑模型的ID';
COMMENT ON COLUMN "public"."building"."buildingdata" IS '3D建筑模型的geojosn数据';
COMMENT ON COLUMN "public"."building"."buildingclass" IS '类别';
COMMENT ON COLUMN "public"."building"."buildingname" IS '3D建筑模型的名称';
COMMENT ON COLUMN "public"."building"."geometry" IS '面数据';
COMMENT ON COLUMN "public"."building"."create_by" IS '创建人';
COMMENT ON COLUMN "public"."building"."create_time" IS '创建时间';
COMMENT ON COLUMN "public"."building"."update_by" IS '更新人';
COMMENT ON COLUMN "public"."building"."update_time" IS '更新时间';
COMMENT ON COLUMN "public"."building"."processdata" IS 'mvt数据';
-- Indexes structure for table building
CREATE INDEX "buildingid_building_index" ON "public"."building" USING btree (
"buildingid" "pg_catalog"."int8_ops" ASC NULLS LAST
);
CREATE INDEX "geometry_building_index" ON "public"."building" USING gist (
"geometry" "public"."gist_geometry_ops_2d"
);
CREATE INDEX "id_building_index" ON "public"."building" USING btree (
"id" "pg_catalog"."int8_ops" ASC NULLS LAST
);
CREATE INDEX "st_buffer_building_index" ON "public"."building" USING gist (
st_buffer(geometry, 0::double precision) "public"."gist_geometry_ops_2d"
);
-- Primary Key structure for table building
ALTER TABLE "public"."building" ADD CONSTRAINT "building_origin_copy1_pkey" PRIMARY KEY ("id");
What You Expected?
能否调整排除优先级,让exclude_columns先与类型识别生效?
或者
对不支持或无法识别的数据类型统一默认为string,而不是直接失败?
How to Reproduce?
No response
Anything Else?
No response
Are you willing to submit PR?
Code of Conduct